Soliloquy is developing a software application that uses speech recognition technology to provide reading support to children, bringing to market a new opportunity to combine the key benefits of one-on-one support, guided oral reading, and independent reading. The application pulls key elements of fluency instruction into its software features, prompting children to read aloud, tracking student progress, and providing interventions where appropriate as children read. This research study will build on Soliloquy's previous work, conducting child reading speech analysis on a larger scale. Toward the goal of improving the accuracy of our intervention strategies, this research will analyze major specific disfluency behaviors in terms of temporal patterns and errors exhibited by children as they read. The research will also draw implications about leveling, indicated by non-specific disfluency errors During Phase I, Soliloquy will 1) create a software tool by which to collect and examine fluency data; 2) gather speech pattern data from children in the form of audio files; 3) analyze audio files for frequency of three major specific disfluency difficulties - orthographic, vocabulary/concept, and integration difficulties - distinguishing them from non-specific disfluency difficulties (non-specific difficulties, unlike specific difficulties, are related to word recognition rather than reading comprehension); and 4) develop intervention and leveling specifications for the next version of the product according to the findings of the disfluency analysis, to be developed in Phase II.